How to Storyboard When You Cannot Draw

Key Takeaways

• A storyboard communicates a shot; it is not a portfolio piece. Legibility beats craft every time.
• Four things per panel: framing, subject position, camera position, and what changes. Nothing else is required.
• Reference photographs are the fastest non-drawing method and often clearer than a sketch.
• Stick figures in a labelled box genuinely work and take thirty seconds a panel.
• Photograph your own staging with a phone. It is the most underused method and the most accurate.
• Only board the shots that need it. Most sequences do not, and a full board is usually wasted effort.

What a Panel Has to Do

Before choosing a method, be clear about the job, because it is much smaller than people assume.

Framing. Wide, medium, close. This is the single most important piece of information and it can be conveyed by the proportion of the frame the subject occupies. No drawing skill involved.

Subject position. Where in the frame, facing which way. A box with a labelled shape does this.

Camera position and height. Where the camera is relative to the subject, and whether it is above, at, or below eye level. Often better expressed in a one-line note than in a drawing, and frequently clearer as a simple overhead diagram.

What changes. Whether anything moves, and what. An arrow plus three words.

That is the entire required payload. Everything else on a beautifully drawn board (lighting mood, expression, texture, costume detail) belongs to the lookbook and the moodboard, not here. Boards that try to carry it become expensive to produce and slower to read.

A useful diagnostic: show a panel to someone and ask what size the shot is and where the camera is. If they can answer, the panel works, regardless of how it looks.

Five Methods That Do Not Need Drawing

1. Reference photographs

Find a frame from an existing film, photograph or advert that matches the shot you want, and use it as the panel with a one-line note saying what to take from it.

The fastest method and often the clearest, because a photograph carries framing and camera position unambiguously, whereas a sketch has to imply them.

The one risk is over-reading. A reference contains wardrobe, location and lighting that you are not proposing. Always annotate what to extract: "the framing and the camera height, ignore everything else". Without that line, someone will match the wrong thing.

2. Your own staged photographs

Set the scene with whoever is nearby, hold up a phone, and take the picture.

This is the most underused method and probably the best. It gives you the actual space, the actual proportions, and a frame you know is achievable, which a found reference does not. It also surfaces problems early: you discover on the recce that the shot you imagined does not fit the room, which is much cheaper than discovering it on the day.

Do it at the location if you can, and use the same lens equivalent you intend to shoot on so the framing is honest.

3. Boxes and stick figures

Draw a rectangle in the aspect ratio. Put a stick figure where the subject goes, sized to indicate the shot size. Arrow for movement. Label anything ambiguous.

This looks crude and communicates precisely. A first AD does not need a rendering; they need to know it is a two-shot at waist height with a push in. Thirty seconds a panel and no skill required.

The commonest objection is that it looks unprofessional in front of a client. The answer is that clients see the lookbook and style frames; the storyboard is a crew document.

4. AI-generated frames

Generate panels from a text description of the shot.

Genuinely useful now, with two cautions. It is easy to generate something beautiful and unachievable, which sets an expectation the shoot cannot meet. And AI frames are cheap enough to make in volume, which tempts people to board everything, including the shots that did not need boarding.

Use it after the direction is locked, generate from your own lookbook rather than from imagination, and check every frame against what you can actually shoot.

5. Existing footage and screen grabs

For anything with an established look, or a reshoot, or a series, grab frames from what already exists. Unambiguous and instantly legible to a returning crew.

Whichever method you use, the panels need to sit in sequence with the intent for each shot beside them, and that is a layout problem rather than an art problem. On a Storyflow canvas the panels are objects in a row with the shot intent and the notes attached, which is what a crew actually reads from. Its AI can generate frames on Pro and Max, so a board can mix your staged photographs, found references and generated panels in one sequence rather than forcing a single method.

For productions that need formal boards with panel numbering, scene linking and export to a shooting script, Boords and StudioBinder are the specialists and this does not replace them.

Board Less Than You Think

The most useful advice here is not about method.

Most sequences do not need a storyboard. Coverage of two people talking in a room is standard grammar; boarding it produces panels nobody consults. What genuinely benefits from boarding is a short list.

Board these: anything with visual effects, anything where the geometry is complex or could confuse the audience, action, anything expensive enough that a wrong decision on the day is costly, anything a client has to approve in advance, and any sequence where several departments have to be coordinated to the same frame.

Do not board: ordinary dialogue coverage, documentary and anything unscripted, and sequences you are deliberately discovering on the day.

A shot list plus three or four boarded sequences is the right shape for most productions and takes a fraction of the effort of a full board, while capturing nearly all the benefit.

Practical Assembly

Work in sequence, not in order of difficulty. Boarding the interesting shots first and filling in later produces gaps you discover at the end.

Number panels to match the shot list. The two documents get used together and mismatched numbering is a genuine source of confusion on set.

Add one line of intent per panel. What the shot is for, not what it shows. This is what lets a crew adapt when the location does not cooperate.

Show it to someone cold. Hand it to a person who has not been in the conversation and ask them to describe three shots. Whatever they get wrong is what the board is failing to communicate, and it is usually camera position.

What Goes Wrong

Trying to make it beautiful. Time goes into rendering rather than communicating, and boards get abandoned halfway.

References used without annotation. The crew matches the wardrobe instead of the framing.

AI frames generated before the look is locked. Sets an expectation the shoot cannot meet.

Boarding everything. Weeks of effort on panels nobody opens.

No camera position. The commonest omission and the thing cold readers most often get wrong.

Panel numbers that do not match the shot list. Confusion on the day, every time.

The Bottom Line

Storyboarding is a communication task, not a drawing task. Photograph your own staging, use annotated references, draw boxes with stick figures, or generate frames once the look is locked. All four work, and the staged phone photograph is probably the best and least used.

Include the framing, the subject position, the camera position and what changes, add a line of intent, and board only the sequences that genuinely need it. A crude board that answers those questions beats a beautiful one that does not.